Ansaro

Sam Stone's frustration with hiring's subjectivity planted the seed for Ansaro in 2011. As a recent college graduate at Bain, he was assigned to review Stanford student applications alongside peers, and struck by how wildly their assessments diverged. "We were aligned on the outcomes that defined a good hire (high-performance ratings, retention), but we disagreed on the inputs that generated those outcomes," he recalls. Some colleagues swore STEM backgrounds were essential; others championed humanities majors. Some valued club leadership; others fixated on GPA. Stone realized these questions could be answered analytically—by mapping performance records back to job applications and hunting for correlations.

He began exploring the problem in his spare time but quickly hit three barriers: data was trapped in homegrown systems without APIs, HR leaders glossed over when he mentioned analytics, and outcome data itself could be biased. Stone paused. But by 2016, while finishing his MBA at Harvard, the landscape had shifted. Cloud-based HR systems with API capabilities were becoming standard, and HR leaders were growing excited about data science. When he asked companies to share anonymized employee data, a few agreed. Stone confirmed the signal: these companies' application data had genuine predictive power for performance and retention—insights they weren't yet using. Armed with proof-of-concept, he went full-time immediately after graduating.

Ansaro's first iteration was bespoke. For their initial handful of enterprise customers, Stone and his team built customized predictive hiring models from scratch. This approach had clear advantages—they made no assumptions and could tailor models to each customer's data and questions, generating impressive presentations and discussions. But it had a fatal flaw: none of these models were production-ready. "With each new customer, we essentially started over," Stone admits. After repeating this cycle, they pivoted to building a scalable platform that could ingest data, fit models, and return suggestions about live applicants across multiple customers. The plan crumbled on execution: none of their customers would grant live access to their HR systems. Some lacked APIs; others cited security concerns or bureaucratic friction. Meanwhile, Stone realized customers didn't even trust their own recruiting data—applicants game systems and lie on applications.

That realization triggered a second pivot toward interviews. Most customers believed structured interviews, when done well, surfaced more signal than applications or resumes, yet interviews were often conducted poorly. Stone's team built a platform for companies to plan structured interviews, assign panels, and record standardized feedback. The headline feature was an AI notetaker: for phone interviews, Ansaro provided a conference line, automatically recorded calls with participant notification, and sent recruiters AI-generated transcripts and summaries. The product was sophisticated, but it solved the wrong problem in the wrong way.

Sales relied almost entirely on Stone's personalized outreach. He sourced prospects—CHROs and heads of recruiting at enterprises with over 1,000 employees—primarily through LinkedIn, then used Hunter.io to infer email addresses and ran drip campaigns via HubSpot. His cold emails carried personalization and generated a ~3% positive response rate. Stone traveled when prospects were large or sufficiently engaged. Secondary channels included conference attendance and their Stanford/Harvard alumni networks, though those dried up quickly. The team experimented with content marketing—blog posts on HR, data science, and better interviewing—and sponsored "HR + Data Science Discussion and Dinner" events with guest speakers and a mix of customers and prospects. They avoided paid marketing entirely.

The company landed "a number of large enterprise customers," but they never graduated beyond pilot contracts. Total revenue across two years reached only $100,000, nearly all from non-recurring, services-heavy pilots—the opposite of the recurring SaaS model they'd envisioned.

Nothing truly worked, but the failures were instructive. The second-pivot interview platform ran into two insurmountable problems. First, companies already prioritizing interview improvement typically used modern applicant tracking systems like Lever or Greenhouse, which offered all of Ansaro's functionality plus more—the only missing piece was the AI notetaker. Second, companies that ignored interview quality weren't interested in buying tools to improve it. The AI notetaker itself—the marquee feature—didn't solve a meaningful pain point. Notetaking is annoying, not a billion-dollar problem, and the transcription quality was mediocre enough that recruiters preferred spending their own time taking notes over editing faulty transcripts.

Stone identifies three systemic failures. First, the team was too slow to pivot because their culture discouraged challenging foundational product assumptions. "When our initial idea turned out to be bad, it took too many months for others to question it and for me to abandon it." Second, they conflated buyers with users. CHROs spoke passionately about hiring better, but recruiters—the actual users—cared more about efficiency than quality. Ansaro pitched new hire quality to executives but never resonated with the people using the tool. Third, and perhaps most damning, they tackled a problem requiring years to measure. Determining whether a hire was successful demands months or years of observation. "That's way too long for a small startup," Stone reflects. Backtesting arguments never convinced HR buyers.

Ansaro shut down at the end of 2018, two years after launch. Immediately before closure, the team of six burned $60-70K monthly. Against that, they'd raised $3 million—$2.25 million from institutional investors led by Silicon Valley Data Capital and $750,000 from friends and family. Stone pitched roughly 30 investors before securing that term sheet. He spent $100K in total revenue and returned remaining capital to investors.

Stone now works as a product manager at Opendoor, building machine-learning products for real estate. Looking back, he wishes he'd waited for genuine product-market fit and recurring software users before raising institutional capital. He'd have raised only from friends and family. For future founders, he prescribes building products easily and quickly tested—something SMBs or individuals can trial without requiring enterprise-wide commitment, with measurable benefit within hours or days, not months or years. He also warns against relying on personal charisma for initial sales, which doesn't scale and resists experimentation. Scalable channels—paid marketing, SEO, partnerships—matter more.